Evaluation of Long-Term Prestress Losses in Post-Tensioned Box-Girder Bridges

摘要

Most of the recent highway bridges built in California have post-tensioned, cast-in-place, concrete box-girder superstructures rigidly connected to bridge columns. However, methods provided in the current (2007 and 2010) AASHTO LRFD Bridge Design Specifications for estimating long-term prestress losses are essentially based on research focused on pretensioned members and are not adequate for post-tensioned bridge girders. Long-term prestress losses in post-tensioned members are expected to be smaller than those in pretensioned members due to two main factors. One is the higher amount of mild reinforcement present in posttensioned bridge girders, which provides a higher restraint to the creep and shrinkage of concrete, and the other is that post-tensioning could take place a long while after the girders have been cast and the concrete has reached a more mature age, which results in a lower creep. The main objectives of the study reported here were to assess the accuracy of the long-term prestress-loss estimation methods given in the current AASHTO LRFD Specifications for post-tensioned bridge girders, and to develop more suitable analysis methods for these members. One is a refined analysis method and the other is a simplified method also referred to as the approximate analysis method. Both methods have been validated and proven to be accurate using field data collected from two bridge structures that were monitored for prestress losses over a period of more than four years. Furthermore, formulas recommended in the AASHTO 2004 and 2007 Specifications for calculating the creep and shrinkage of concrete have been evaluated with the material data obtained from concrete cylinders cast with the same batches of concrete used for the monitored bridge girders, and have been used for the loss calculations. It has been found that the formulas in AASHTO 2004 provide a much better correlation with the creep and shrinkage data obtained from the concrete cylinders than AASHTO 2007. In general, the AASHTO 2007 formulas significantly under-estimate the creep and shrinkage of the concrete cylinders, and, thereby, result in calculated long-term losses that are much lower than the measured values. Hence, the creep and shrinkage formulas provided in AASHTO 2004 are recommended for the prestress-loss calculations using the proposed refined analysis method. Finally, suggestions are provided for possible implementation of the proposed analysis methods in the AASHTO LRFD Specifications for calculating long-term prestress losses in post-tensioned bridge girders. Both methods are expressed in forms that can be readily implemented in the AASHTO LRFD Specifications.